Bug 93535 – [chromium/mac] Map NSEventPhaseMayBegin appropriately on 10.8

• Component: New Bugs
• Status: RESOLVED FIXED

[chromium/mac] Map NSEventPhaseMayBegin appropriately on 10.8

Created attachment 157307 [details]
Patch

Nico: Please review the change.
Dimitri: Will need you approval (assuming Nico still isn't a reviewer).

[Nico Weber]

Comment on attachment 157307 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=157307&action=review

LGTM, thanks for doing this. (Bug 79868 has some background.)

> Source/WebKit/chromium/src/mac/WebInputEventFactory.mm:57
> +#if __MAC_OS_X_VERSION_MAX_ALLOWED < 1080

https://developer.apple.com/library/mac/#documentation/Cocoa/Reference/ApplicationKit/Classes/nsevent_Class/Reference/Reference.html doesn't mention this for 10.8 either , so maybe omit the #if completely?

Comment on attachment 157307 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=157307&action=review

>> Source/WebKit/chromium/src/mac/WebInputEventFactory.mm:57
>> +#if __MAC_OS_X_VERSION_MAX_ALLOWED < 1080
> 
> https://developer.apple.com/library/mac/#documentation/Cocoa/Reference/ApplicationKit/Classes/nsevent_Class/Reference/Reference.html doesn't mention this for 10.8 either , so maybe omit the #if completely?

It's in the 10.8 SDK NSEvent.h header though...

Wouldn't omitting it result in a build error when building against the 10.8 SDK? (Or do compilers ignore multiply-defined enums with same values these days like they do with typedefs?)

[Nico Weber]

(In reply to comment #4)
> (From update of attachment 157307 [details])
> View in context: https://bugs.webkit.org/attachment.cgi?id=157307&action=review
> 
> >> Source/WebKit/chromium/src/mac/WebInputEventFactory.mm:57
> >> +#if __MAC_OS_X_VERSION_MAX_ALLOWED < 1080
> > 
> > https://developer.apple.com/library/mac/#documentation/Cocoa/Reference/ApplicationKit/Classes/nsevent_Class/Reference/Reference.html doesn't mention this for 10.8 either , so maybe omit the #if completely?
> 
> It's in the 10.8 SDK NSEvent.h header though...

Ah ok, then it's correct as is (I don't have a 10.8 SDK at hand to check).

> 
> Wouldn't omitting it result in a build error when building against the 10.8 SDK? (Or do compilers ignore multiply-defined enums with same values these days like they do with typedefs?)

No, multiply-defined enums are an error. I assumed the documentation would match the header. Silly me.

[Dimitri Glazkov (Google)]

Comment on attachment 157307 [details]
Patch

rs=me

[WebKit Review Bot]

Comment on attachment 157307 [details]
Patch

Clearing flags on attachment: 157307

Committed r125125: <http://trac.webkit.org/changeset/125125>

[WebKit Review Bot]

All reviewed patches have been landed.  Closing bug.

For historical purposes, this fixed http://crbug.com/139553.